Modeling probabilities of patent oppositions in a Bayesian semiparametric regression framework

Previous econometric analyses of patent data rely on regression methods using purely parametric forms of the predictor for modeling the dependence of the response. These approaches lack the capability of identifying potential non-linear relationships between dependent and independent variables. In this paper, we present a Bayesian semiparametric approach making use of Markov Chain Monte Carlo (MCMC) simulation techniques which is able to capture these non-linearities. Using this methodology we reanalyze the determinants of patent oppositions in Europe for biotechnology/pharmaceutical and semiconductor/computer software patents. Our semiparametric specification clearly finds considerable non-linearities in the effect of various metrical covariates which has been not been discussed previously. Further, a formal model validation based on ROC-methodology which splits the data in a training and a validation data set shows a significant improvement of the explanatory and the predictive power of our approach compared to purely parametric specifications.